(1) Field of the Invention
The invention relates to the general field of Liquid Crystal Displays, particularly color balance in the emergent light.
(2) Description of the Prior Art
Referring to FIG. 1, the basic parts of a liquid crystal display are schematically illustrated in cross-section. A number of layers are involved, the outermost being a pair of crossed polarizers 1 and 2. In their most commonly used configuration, the polarizers are arranged so as to have their optic axes orthogonal to one another. That is, in the absence of anything else between them, light passing through the entrance polarizer would be blocked by the exit polarizer, and vice versa.
Below the entrance polarizer 1 is an upper transparent insulating substrate 3 (usually glass) and immediately above the exit polarizer 2 is a similar lower substrate 4 . Conducting lines, such as 6, running orthogonal to, and insulated from, one another are located on the upper (inward-looking) surface of 4. Said orthogonal lines are connected at their intersections, optionally through Thin Film Transistors (TFTs). This allows voltage, separately applied to a pair of orthogonal lines, to be added together only at the intersecting position which will overlie a given pixel (or sub-pixel) of the display.
Sandwiched between, and confined there by means of suitable enclosing walls (not shown), is a layer of liquid crystal 5. Liquid crystals comprise long thread-like molecules whose orientation, relative to a given surface can be controlled by coating such a surface with a suitable orientation layer (not shown) and rubbing said orientation layer in the desired direction just prior to bringing it into contact with the liquid crystals.
Thus, in FIG. 1, the molecules closest to upper substrate 3 might be oriented so as to lie in the plane of the figure while the molecules closest to lower substrate 4 would be oriented to lie perpendicular to this plane. Molecules in between the two sets of pre-oriented molecules then arrange themselves so as to gradually change their orientation between these two extremes. Hence the term `twisted nematic` (TN) for such a configuration. A TN is optically active and will rotate the plane of any polarized light that traverses it.
Thus, polarized light that was formed and oriented as a result of passing through entrance polarizer 1 will be rotated though an angle (for example 90.degree.) after traversing layer 5 and so will be correctly oriented to pass through the exit polarizer 2. Such a device is therefore normally on (transmits light).
An important property of TN is that, in the presence of an electric field (typically about 1,000 volts/cm.), normal to the molecules, said molecules will all orient themselves so as to point in the same direction and the liquid crystal layer will cease to rotate the plane of polarization. As discussed above, a single pair of orthogonal lines comprise one electrode for generating said electric field, the other being transparent conducting common electrode 10, comprising indium-tin-oxide (ITO).
Besides the TN structure, there is another possible configuration wherein all molecules of liquid crystal 5 (in FIG. 1) are oriented to lie in the plane of the figure but the inclination of each is different. The molecules closest to upper substrate 3 might be inclined `right up and left down` while the molecules closest to lower substrate 4 would be inclined `left up and right down`. These inclination angles (7.degree. in our example) can be controlled by selecting the orientation layer. Molecules in the middle position between upper and lower substrates might be perpendicular in the presence of the electric field. Other molecules, between these three sets of molecules, then arrange themselves so as to gradually change their orientation (bending) between the three boundaries. Hence the term `bend cell` for such a configuration. Examples of bend cells are described by C. L. Kuo et al. in Appl. Phys. Letters vol. 68 March 1996 pp. 1461-1463, and by T. Miyashita et al. in Jpn. Jour. Appl. Phys. vol. 34 February 1995 pp. L177-L179.
In addition to being capable of rotating the plane of polarization, as in a TN structure, liquid crystals are also birefringent. This means that for a plane polarized wave of light there are different refractive indices for the two components of the electric vector. As a result, after passing through a given thickness of a bi-refringent layer, the phase difference between these components (normally 0) changes, resulting in an elliptically polarized wave.
After passing through the exit polarizer 2, said elliptically polarized light is converted once more to plane polarized light. However, its intensity will have been changed, depending on the value of the afore-mentioned phase change which can be modified by varying the voltage applied across the liquid crystal molecules.
To view a display of the type illustrated in FIG. 1, light may be applied from above the entrance polarizer, and viewed from below the exit polarizer or a reflecting surface may be applied to the lower surface of exit 2 polarizer and the device viewed from above.
In general, color LCDs are built in the same way as monochrome LCDs except that their light has been passed through color filters. The latter comprise a matrix of sub-pixel size regions, such as 7, 8, or 9 in FIG. 1, on common substrate 3, each region being a tiny single filter. The spatial locations of the different colored regions are known to the liquid crystal display control system which determines the amount of light that is allowed to pass beyond any given dot, thereby creating a color image. For example, in FIG. 1, region 7 might represent a green filter, region 9 a blue filter, and region 8 a red filter.
One of the ways in which multicolor filters are manufactured is by using a light sensitive resin (such as a methacrylate polymer) as the material out of which the aforementioned dots are formed. Such a resin can be made to serve as a light filtering medium by dispersing an appropriate pigment within it. Then, by using a mask when exposing such a resin to the appropriate actinic radiation, any desired pattern of sub-pixel-sized regions of a given color can be produced.
A commonly incorporated feature of LCDs is a black matrix, (not shown in FIG. 1). Its purpose is to block light that reaches layer 2 without having passed through the open space between color filters. Such light is extraneous to the display and reduces the overall contrast.
Besides changing the overall brightness of the display, there is a more serious problem associated with the birefringent phenomenon, as described above. The magnitude of the phase change varies with wavelength so that the relative intensities of the emerging colors differ from that of the original image. This dispersive effect is particularly noticeable with respect to light that was intended to appear white but, instead, appears slightly colored.
A partial solution to this problem has been described in U.S. Pat. No. 5,136,405 (Wada et al. August 1992) and U.S. Pat. No. 5,380,459 (Kanemoto et al. January 1995). These patents teach the use of a single (macro) compensator comprising a sheet of bi-refringent material (11 in FIG. 1) that introduces additional phase changes, for the purpose of cancelling out those introduced by the liquid crystal. Unfortunately, the dispersion of the macrocompensator material cannot be matched exactly to that of the liquid crystal so exact cancellation of the color distortion introduced by the liquid crystal is not possible with a macro-compensator and a conventionally designed liquid crystal display.
An attempt to match the dispersion of the liquid crystal layer to that of a macrocompensator has been described by Haim et al. in U.S. Patent 5,402,141 (March 1995). Haim's approach is to modify the dimensions of the cell gaps within the display itself so that the phase changes that are introduced are better suited to correction by a macrocompensator. As will be seen, this approach is quite different from that of the present invention.